Service disruptions in a packet network, e.g., a VoIP network can occur as a result of mass calling events that produce overloads and subsequent queuing congestion of signaling messages that result in call blocking due to call set up signaling message processing delays. If the call volumes quickly subside then the network typically recovers and will soon begin processing calls at a normal rate. However, if the call volumes decline and return to normal but the corresponding call blocking rate continues to increase or even remain flat, then the network is suffering a congestion problem that requires corrective actions to be taken to flush out the congested signaling message queues before calls can be handled normally after the call volume rate returns to normal.
Therefore, a need exists for a method and apparatus for automating the detection and clearance of congestion in a packet network, e.g., a VoIP network.